This application is in response to FA 88-CA-02: National Collaborative Diagnostic Imaging Trail Projects. Rapid advances in medical imaging have occurred in the past 10 years with associated increases in the cost of imaging procedures. In view of the many available modalities, clinically efficient as well as cost effective diagnostic imaging strategies need to be developed. The long term objective of this proposal is to evaluate the relative roles of the relevant imaging modalities in the diagnosis and staging of colorectal and pancreatic cancer. Specifically, the relative accuracy of computed tomography (CT) and magnetic resonance imaging (MRI) in the detection of liver metastases will be determined. The accuracy of gastrointestinal barium studies, CT, MRI, ultrasonography and angiography will be evaluated in their ability to stage the loco-regional extent of colorectal and pancreatic cancer. Using receiver operating characteristic curves, we will compare the performance of each modality separately and then jointly in the diagnosis of the specific relevant abnormalities and attempt to determine appropriate diagnostic criteria for staging colorectal and pancreatic cancer and optimal use of each modality. The role of CT, and MRI in the monitoring of patients with colorectal cancer will also be tested through sequential examinations within the first 24 months after initial therapy. Detailed pathologic correlations are proposed as measures of truth for both loco-regional extent and hepatic metastases. The data acquired through this prospective study will form the basis for developing optimized decision trees in imaging of colorectal and pancreatic cancer.